Binary outcome in instrumental variable analysis

Ro8

New Member
#1
I'm looking to perform an instrumental variable analysis using the following variables:
I. a continuous endogenous variable (exposure levels)
II. a continuous instrument variable for (I.)
III. two exogenous variables (one binary, one continuous)
IV. a binary outcome (outcome y/n).

However with my limited statistical background i'm having trouble deciphering the background information of a few Stata commands (and thus deciding which to use). As i currently understand it, i could use either qvf (to determine a ratio between the logistic regression of IV on I with the linear regression of II on I) or ivprobit (using either a two-step or maximum-likelihood estimator). Are there any other commands i should be aware of, and is my current understanding of qvf and ivprobit correct?

Thank you in advance for any help.